DESCRIPTION Enterocolitis caused by enterohemorrhagic Escherichia coli (EHEC), i.e., E. coli that produce Shiga-like toxins (SLTs), usually precedes HUS, but fewer than 10% of children with EHEC infection progress to HUS. We hypothesize that individual differences in toxin receptors of the host account for some of this differential susceptibility to HUS, and that these receptors are essential determinants of HUS pathogenesis. Following EHEC infection, SLT may enter the circulation and bind to host cell globotriaosylceramide (Gb3) toxin receptors; such binding in the endothelium is thought to be central to HUS pathology. Globotriaosylceramide is one of the major glycolipids that binds to SLTs in solid-phase assays, and Gb3 is accepted as a major functional receptor for SLT toxicity to cultured cells. Different molecular species of Gb3 that differ in their specific fatty acid moieties have different binding affinities to SLT. Variation of specific species of glycolipids among tissues is not known; such variations may underlie SLT sensitivity. The relative affinities of individual Gb3s for SLT differ greatly under different assay conditions; therefore, large, well defined and carefully studied human populations will be used to relate individual levels of specific species of Gb3 with susceptibility to HUS. We propose that an underlying genetically based predisposition toward HUS will be revealed by erythrocyte glycolipid analyses. The relationship between incidence of HUS, blood group type, and erythrocyte Gb3 profiles will be studied retrospectively in a genetically homogeneous population with unusually high HUS susceptibility, and prospectively in EHEC-infected Canadian children, with emphasis on defining the roles of different species of Gb3 in the etiology of HUS. The relationship between endothelial Gb3 profiles and erythrocyte Gb3 profiles will be studied in blood and human endothelial cells derived from the same umbilical cords. The relationship between endothelial cell Gb3 and susceptibility to SLT will be studied in cultured human endothelial cells. Thus, this project addresses the role of specific toxin receptors in the pathobiology of HUS, the relationship between regulation of host cell receptors and HUS susceptibility, and genetic differences in human glycolipid metabolism that relate to susceptibility to vascular endothelial injury. Such information may identify aspects of HUS pathobiology that are amenable to therapeutic intervention and may aid in identifying patients at high risk for HUS who could be targeted for increased vigilance and prophylactic measures.